JP7014018B2 - Control system - Google Patents

Description

The present disclosure relates to remote control of self-driving cars.

Patent Document 1 discloses that when an autonomous vehicle enters a remote control target area, an operator remotely controls the autonomous vehicle by using a remote control device.

Japanese Unexamined Patent Publication No. 2017-147626

In the case of the above prior art, how to assign an operator to an autonomous vehicle that requires remote control is not considered. Based on the above, the present disclosure aims to assign an operator to an autonomous vehicle so that a smooth traffic flow can be realized.

One embodiment of the present disclosure comprises a communication device (24) that communicates with a plurality of autonomous vehicles (800) equipped with a communication unit (830), and an interface (120) by an operator for remote control of the autonomous vehicle. A control device (20) that receives a command input to the vehicle and remotely controls the autonomous vehicle using the communication device; among the plurality of autonomous vehicles, the autonomous vehicle to be remotely controlled. A specific unit (S221) that identifies a target vehicle; and a calculation unit (S231, S232, S252, S260) that calculates the work time and priority required for remote control of the target vehicle based on the traffic conditions around the target vehicle. ) And; a determination unit (S272) that determines the processing order, which is the order in which remote control is executed for the target vehicle, using the calculation result by the calculation unit; according to the processing order determined by the determination unit. It is a control device including an allocation unit (S340) that allocates remote control work to the operator.

According to this form, the work time and priority required for remote control are calculated, the processing order is determined based on the work time and priority, and the remote control work is assigned to the operator according to the processing order, so that the operation is smooth. A good traffic flow is realized.

The block diagram which shows the internal structure of a control device and a remote control device. A block diagram showing the internal configuration of an autonomous vehicle. A flowchart showing the allocation process. A flowchart showing a processing order determination process. A map containing areas that require remote control. The figure which shows the traffic condition in the area X. The figure which shows the traffic condition in the area Y. The figure which shows the traffic condition in the area Z. A sequence chart showing the state of remote control. The figure which shows the division of roles of an autonomous vehicle, a control device, and an operator. A flowchart showing a processing order determination process. The figure which shows the assignment of work by an operator. A flowchart showing a processing order determination process. The figure which shows the assignment of work by an operator. The figure which shows the state of the cost calculation for deciding the allocation of work. A flowchart showing the allocation process. The figure which shows the state of the remote control in the area X. The figure which shows the state of the remote control in the area X. The figure which shows the state of the remote control in the area X.

As shown in FIG. 1, the control device 20 includes a CPU 22, a communication device 24, and a storage device 26. The communication device 24 executes wireless communication with a plurality of autonomous vehicles 800 and a plurality of manually driven vehicles 900, and wired communication with a plurality of remote control devices 100. The communication with the remote control device 100 may be wireless communication. The storage device 26 includes a non-transitional execution storage medium such as a semiconductor memory. The storage device 26 stores a map database 28 and a program 29 for executing various processes described later.

The remote control device 100 includes a communication device 110 and an interface 120. The communication device 110 executes communication with the control device 20. The interface 120 is a man-machine interface for input / output. Specifically, the information collected by the automatic driving vehicle 800 and the manual driving vehicle 900 acquired via the communication device 110 is notified to the operator, and the operation input by the operator is acquired and via the communication device 110. It is transmitted to the control device 20. In the present embodiment, a predetermined number (for example, 10) of remote control devices 100 are prepared.

As shown in FIG. 2, the self-driving car 800 includes sensors 810, a self-position estimation unit 820, a communication unit 830, a drive unit 840, and a control unit 850.

The sensors 810 include an inner world sensor and an outer world sensor. The internal sensor acquires information such as the traveling speed and steering angle of the own vehicle. The external sensor is composed of a camera, LIDAR, millimeter wave radar, ultrasonic sonar, etc., and collects information around the vehicle. Hereinafter, the information collected by the sensors 810 is also referred to as sensing data.

The self-position estimation unit 820 estimates the self-position of the self-driving car 800. Specifically, the self-position estimation unit 820 uses both self-position positioning using GNSS and odometry. Odometry is a method of estimating a self-position using a gyro sensor, an acceleration sensor, or the like included in the self-position estimation unit 820. In other embodiments, either self-positioning using GNSS or odometry may be used. The communication unit 830 communicates with the communication device 24 included in the control device 20. The communication unit 830 transmits the sensing data to the control device 20. Further, the communication unit 830 receives the information for remote control transmitted from the control device 20.

The drive unit 840 is composed of an acceleration mechanism, a deceleration mechanism, a steering mechanism, and the like, and realizes the running of the autonomous driving vehicle 800.

The control unit 850 is composed of one or a plurality of ECUs. The control unit 850 includes a storage device 851. The storage device 851 stores the ID 852 and the map database 853. ID 852 is information for the control device 20 to identify the self-driving car. Map information is stored in the map database 853.

The control unit 850 realizes a process for realizing level 4 automatic operation by controlling the drive unit 840 using the information acquired from the sensors 810 and the self-position estimation unit 820. Specifically, by determining a travel plan from the current location and the destination and acquiring information such as a signal and a preceding vehicle, acceleration / deceleration and steering are automatically executed using the drive unit 840. Therefore, the self-driving car 800 can run even if the driver is not on board.

Further, the control unit 850 repeatedly transmits the sensing data, the position information indicating the current location, the travel plan, the ID 852, and the travel data to the control device 20 using the communication unit 830. Hereinafter, the information transmitted in this way is collectively referred to as vehicle information. The traveling data is a general term for data related to traveling such as a traveling locus and a vehicle speed obtained from the data collected by the sensors 810 and the self-position estimation unit 820.

Further, when the control unit 850 receives the remote control instruction via the communication unit 830, the control unit 850 controls the drive unit 840 according to the instruction.

The manual driving vehicle 900 has the same configuration as the automatic driving vehicle 800. Therefore, the manually driven vehicle 900 transmits the vehicle information to the control device 20 by the communication unit. However, the manually driven vehicle 900 does not execute autonomous driving or remote controlled driving. That is, the manually driven vehicle 900 is driven by the driver. It should be noted that the manually driven vehicle 900 only needs to be able to grasp the position information and the vehicle speed, so that not all the configurations included in the sensors 810 are indispensable. The above-mentioned "similar configuration" means having at least a configuration for acquiring position information and vehicle speed and transmitting them to the control device 20.

The CPU 22 repeatedly executes the allocation process shown in FIG. When the CPU 22 starts the allocation process, the CPU 22 executes the process order determination process as S200. As shown in FIG. 4, first, the CPU 22 analyzes the vehicle information acquired from the autonomous driving vehicle 800 and the manually driven vehicle 900 as S211. Subsequently, the CPU 22 identifies, as S221, the autonomous driving vehicle 800 that requires remote control based on the information analyzed in S211. The term "remote control" in the present embodiment includes remote monitoring. The remote control is to remotely drive the self-driving car 800 by giving an appropriate instruction to the self-driving car 800 in a situation where the self-driving car 800 cannot run autonomously. In remote monitoring, the operator monitors the presence or absence of danger based on vehicle information.

For example, in the area X shown in FIG. 5, when traffic control by hand signals is executed as shown in FIG. 6, the self-driving cars 800A and 800E without a driver require remote control. The manual driving vehicles 900B, 900C, and 900D cannot move forward as long as the automatic driving vehicle 800A is stopped. Similarly, the manual driving vehicles 900F and 900G cannot move forward as long as the automatic driving vehicle 800E is stopped.

Further, in the area Y shown in FIG. 5, as shown in FIG. 7, when the road is blocked by an obstacle such as a broken vehicle, the self-driving car 800H without a driver requires remote control. ..

Further, in the area Z shown in FIG. 5, as shown in FIG. 8, when four self-driving cars 800I, 800J, 800K, 800L stop at a crossroad without a signal, the normal traffic rule is applied. Remote control is required when the intersection cannot be entered. In the case shown in FIG. 8, the self-driving cars 800I, 800J, and 800L without the driver are going straight, and the self-driving car 800K is going to turn right. Therefore, if there is no signal, the normal traffic rule (for example, left priority). (Article 36 of the Road Traffic Act)), none of the self-driving cars 800I, 800J, 800K, and 800L can enter the intersection.

The CPU 22 identifies the self-driving car 800 that requires remote control as described above by analyzing the stop time of the self-driving car 800, the analysis of the image pickup result by the camera as the sensors 810 of the self-driving car 800, and the like. In another embodiment, when the self-driving car 800 cannot run autonomously, the self-driving car 800 may request remote support from the control device 20.

Next, the CPU 22 proceeds to S231 and calculates the working time required for remote control. The work time required for remote control is calculated based on the traffic conditions around the travel route of the autonomous vehicle 800 that requires remote control and the information stored in S483 of FIG. 9 to be described later. That is, for example, the CPU 22 searches for information on a similar situation from the information stored in S483 when the vehicle is stopped due to the inability to autonomously travel in a section where traffic regulation is implemented due to construction work or the like. , Calculate the working time based on the information.

Next, as S260, the CPU 22 calculates the priority reflecting the operation policy for each of the autonomous driving vehicles 800 that require remote control.

Next, as S271, the CPU 22 determines the processing order of the remote control based on the priority and the working time reflecting the operation policy, and finishes the processing order determination process. In this embodiment, it is adopted as an operation policy to minimize the degree of influence on the traffic flow.

In the present embodiment, the degree of influence on the traffic flow is evaluated in consideration of the following three parameters. The first is the number of vehicles (hereinafter referred to as standby vehicles) that cannot actually travel due to the self-driving vehicle 800 waiting for remote control. The greater the number of waiting vehicles, the greater the impact on traffic flow. The vehicle referred to here includes an autonomous driving vehicle 800 and a manually driven vehicle 900. The CPU 22 grasps the number of standby vehicles based on the traveling speed and the position information included in the vehicle information collected from the autonomous driving vehicle 800 and the manually driven vehicle 900. Weighting may be performed to calculate the degree of influence on the traffic flow depending on whether the standby vehicle is the self-driving vehicle 800 or the manually driven vehicle 900.

The second is the time that the waiting vehicle has waited so far. The longer this time is, the greater the impact on traffic flow is considered.

The third is to evaluate by subtracting the number of vehicles flowing out from the road from the number of vehicles flowing into the road on which the self-driving car 800 requiring remote control is located. The larger the number, the greater the impact on traffic flow. By evaluating this number, it is possible to evaluate the impact of considering the number of lanes on the road and traffic conditions. For example, when an autonomous vehicle blocks one lane on a road with a plurality of lanes, the manually driven vehicle can detour if the road is open, so that the influence on the surrounding traffic flow is small. On the contrary, if the road is crowded, it can be evaluated that the influence on the traffic flow is large because the manually driven vehicle cannot take evasive action.

In another embodiment, the degree of influence on the traffic flow may be evaluated in consideration of any one or more of the above three parameters.

When the CPU 22 adopts the first parameter and weights the existing traffic flow so as to more consider the influence on the manual driving vehicle 900, the degree of influence on the traffic flow of the automatic driving vehicle 800A is increased. It is evaluated that the highest is 800E, the next is 800I, 800J, 800K, 800L in the same order, and 800H is the lowest. Next, the CPU 22 determines, for example, the processing order of the remote control in the order of 800A, 800E, 800I, 800J, 800K, 800L, 800H based on the evaluation result of the degree of influence on the traffic flow.

When the processing order determination process is completed, the CPU 22 determines, as S310, whether or not there is an autonomous driving vehicle 800 to which an operator is not assigned among the autonomous driving vehicles 800 that require remote control. If the CPU 22 determines YES in S310, the CPU 22 proceeds to S320.

The CPU 22 confirms the allocation status of the operator as S320. For example, if there is an operator whose remote control has ended, the status of that operator is changed from non-assignable to assignable, so confirm such a change.

Next, the CPU 22 determines whether there is an operator that can be assigned as S330. If there is no assignable operator, the CPU 22 determines NO in S330 and returns to S320.

On the other hand, when there is an operator that can be assigned, the CPU 22 determines YES in S330 and proceeds to S340. When the CPU 22 proceeds to S340, the CPU 22 assigns an operator to the self-driving car 800 according to the processing order determined in S200. After that, it returns to S310. If it is determined as NO in S310, the CPU 22 ends the allocation process.

As described above, in the present embodiment, in principle, if there is an autonomous driving vehicle 800 that requires remote control and there is an operator who does not have an allocation, the allocation is executed. Therefore, if there are seven operators that can be assigned, the operators can be assigned to all the autonomous driving vehicles 800 stopped in the areas X, Y, and Z. However, since there are not always seven or more operators that can be assigned, if the number of autonomous vehicles 800 that require remote control is larger than the number of operators that can be worked, it is determined as described above. Operators are assigned according to the processing order.

For example, when remotely controlling the autonomous driving vehicles 800A and 800E stopped in the area X, the operator 1 visually recognizes the hand signal by looking at the captured image of the front acquired from the autonomous driving vehicles 800A and 800E, and automatically drives. Determine the vehicle 800A as the target of remote control.

As shown in FIG. 9, when the remote control for a certain self-driving car 800 is started, until the operator determines that the remote control for the self-driving car 800 should be ended, S410 to S470 is repeated.

First, as S410, the self-driving car 800 transmits the vehicle information to the control device 20. The control device 20 transmits the received vehicle information and the road traffic information to the remote control device 100 (S420). The road traffic information is map data, traffic conditions, etc. in the vicinity of the area where the autonomous vehicle 800, which is the target of remote control, is located. The traffic conditions include, for example, traffic regulation information such as construction sites and sections, traffic congestion information, and the like.

When the remote control device 100 receives the vehicle information and the road traffic information, it displays it on the interface 120 as S430. For example, a captured image in front of the vehicle, a vehicle speed, and the like are displayed, and a map around the vehicle to be controlled is displayed.

The operator 1 visually confirms the information displayed on the interface 120 as S440 to confirm the situation. Subsequently, the operator 1 executes an operation input for remote control through the interface 120 as S450.

Here, the operation input for remote control will be described with reference to FIG. First, the case of autonomous driving will be described. In the case of autonomous driving, the control unit 850 of the autonomous driving vehicle 800 executes all of the traveling plan, determination, waypoint generation, traveling locus generation, and vehicle control. The driving plan is to make a driving plan based on the sensing data. The judgment is to judge whether or not the vehicle can travel according to the travel plan. Waypoint generation is to calculate the points to pass at each time step and generate a waypoint set in order to travel according to the determined travel plan. The travel locus generation is to generate a travel locus that actually travels with the waypoint as a constraint condition. Vehicle control is to control vehicle control (for example, steering, acceleration / deceleration, etc.) when traveling along a determined travel locus.

When performing remote control, the control device 20 instructs at least one of the above steps. Depending on which process is instructed, it is classified into approval type, modification type, instruction / presentation type, and remote control type. In this embodiment, the approval type is adopted. In other embodiments, it may be a modified type, an instruction / presentation type, or a remote control type.

In the approval type, when the traveling plan of the autonomous vehicle 800 deviates from the traffic rules or requires communication with the outside world (for example, a traffic control staff), the operator determines whether or not the driving is possible. In the case of the approval type, the information required for remote driving is the behavior of the autonomous driving vehicle approved by the operator, that is, waypoint information.

In the modified type, the operator corrects the driving plan when the self-driving car cannot make a driving plan or when it is judged that the driving with the drafted driving plan is difficult. For example, instruct "Run 1m to the right, offset" and so on. Upon receiving such an instruction, the self-driving car corrects its waypoint according to the instruction. In the modified type, the information required for remote driving is the behavior instructed by the operator to the autonomous vehicle, that is, the modified waypoint information.

In the instruction / presentation type, the operator instructs the self-driving car as a waypoint with information on the point to be driven. When giving this instruction, the registered driving data can be used, the driving data of the connected car that traveled at the same point immediately before can be used, or the operator can set it based on the sensing data of the remote control target. .. In the case of the instruction / presentation type, the information required for remote travel is the waypoint information input by the operator.

In the remote control type, the operator takes the place of the driver and operates the steering, accelerator, brake, etc. while monitoring the sensing data of the autonomous driving vehicle to be remotely controlled. In the case of the remote control type, the information required for remote travel is the travel locus and the vehicle speed information at each point.

Upon receiving the operation input, the remote control device 100 transmits the remote control instruction information to the control device 20 as S460. Upon receiving the remote control instruction, the control device 20 remotely controls the autonomous vehicle 800 according to the instruction as S470.

By repeating the above steps as necessary, for example, as shown in FIG. 6, when the self-driving car 800A reaches the position of the broken line, the operator 1 sets S475 to indicate that the remote control is terminated by the interface 120. Enter through.

Upon receiving the input to end the remote control, the remote control device 100 notifies the control device 20 that the remote control is to be terminated as S480. Upon receiving the notification that the remote control is terminated, the control device 20 stores the position where the remote control is performed and the working time as S483.

Further, the control device 20 notifies the self-driving car 800 that the remote control is terminated as S485. After that, the control device 20 updates the allocation status of the operator as S490. On the other hand, the self-driving car 800 resumes autonomous driving as S495.

When the remote control of one autonomous vehicle 800 is completed in this way, the control device 20 starts a process for remote control of the autonomous vehicle 800 that requires remote control in the same area. .. For example, in the case of the region X shown in FIG. 6, after the automatic driving vehicle 800A resumes automatic driving, the target of remote control is changed to the automatic driving vehicle 800E.

As S440, the operator 1 confirms whether or not the self-driving car 800E is permitted or prohibited by the hand signal to be followed through the interface 120. When the operator 1 confirms that the self-driving car 800E is permitted to travel, the operator 1 executes the operation input as S450. Then, when the operator 1 determines that the remote control of the autonomous vehicle 800E should be terminated, S475 to S495 are executed. Then, since the self-driving car 800 that requires remote control in the area X does not exist, the operator 1 starts the remote control of the self-driving car 800 stopped in the area Z.

In the case of region Z shown in FIG. 8, for example, if the self-driving cars 800J and 800L are allowed to go straight at the intersection at the same time and then the self-driving car 800I is allowed to go straight at the intersection, remote control is required. Will be resolved. Therefore, the control device 20 sequentially selects the autonomous driving vehicles 800J, 800L, and 800I as remote control targets.

In the case of the region Y shown in FIG. 7, if the autonomous vehicle 800H is remotely controlled so as to drive in the oncoming lane and pass through an obstacle while avoiding a collision with an oncoming vehicle, remote control is required. The situation disappears.

According to this embodiment, the processing order is determined based on the priority and the working time reflecting the operation policy. Therefore, when the number of autonomous vehicles 800 that require remote control is larger than the number of operators who can work, operators can be assigned so that the influence on the traffic flow is small.

The second embodiment will be described. The description of the second embodiment mainly focuses on the differences from the first embodiment. The points not particularly described are the same as those in the first embodiment.

In the present embodiment, as shown in FIG. 11, in the processing order determination process, the CPU 22 executes grouping as S262 after S260. The grouping is performed so that the self-driving cars 800 having a common cause requiring remote control belong to one group. For example, in the region X of FIG. 6, the self-driving cars 800A and 800E are grouped because they have a common cause of requiring remote control. Further, in the region Z of FIG. 8, the self-driving cars 800I, 800J, 800K, and 800L are grouped because they have a common cause of requiring remote control. The common cause of requiring remote control is identified from the position information included in the vehicle information of the autonomous driving vehicle 800 and the manually driven vehicle 900.

Next, the CPU 22 proceeds to S264 and determines the processing order for each group. As a result of considering the above-mentioned three parameters, in the above-mentioned three cases, it is evaluated that the area X has the highest processing order, then the area Z, and then the area Y.

Next, the CPU 22 proceeds to S271 and determines the processing order of the controlled vehicle by reflecting the processing order for each group.

An example in which one operator can be assigned will be described with reference to FIG. 12 when the processing order for each group is reflected. When there is only one operator that can be assigned, the control device 20 assigns operators (specifically, operator 1) in order from the autonomous driving vehicle 800 that is stopped in the area where the processing order is high. That is, as shown in FIG. 12, the control device 20 first causes the operator 1 to take charge of the area X, then the area Z, and finally the area Y.

On the other hand, as shown in FIG. 12, when there are two assignable operators, the control device 20 causes the operator 1 to be in charge of the area X and the operator 2 to be in charge of the area Z. Then, of the operators 1 and 2, the one whose remote control of the area in charge is completed earlier is assigned to the area Y. In the present embodiment, since the remote control ends earlier in the area Z than in the area X, the control device 20 causes the operator 2 to take charge of the area Y.

Further, as shown in FIG. 12, when there are three assignable operators, the control device 20 causes the operator 1 to be in charge of the area X, the operator 2 to be in charge of the area Z, and the operator 3 to be in charge of the area Y. Let me take charge.

According to the present embodiment, operators can be efficiently assigned by using grouping.

The third embodiment will be described. The description of the third embodiment mainly focuses on the differences from the second embodiment. The points not particularly described are the same as those in the second embodiment.

In the third embodiment, the content of the processing order determination process is different from that of the second embodiment. As shown in FIG. 13, S211 and S221 are the same as those of the first and second embodiments.

After finishing S221, the CPU 22 proceeds to S232 and calculates the degree of influence on the traffic flow and the working time required for remote control for the autonomous driving vehicle 800 specified in S221. The degree of influence on the traffic flow and the working time required for remote control are calculated by the same method as in the first embodiment.

Next, the CPU 22 proceeds to S242 and predicts the self-driving car 800 that will require remote control in the near future among the self-driving cars 800 that are transmitting vehicle information. For S242, a travel plan transmitted from the self-driving car 800 is used. That is, by considering the travel plan, it can be predicted that the vehicle will travel in a section that requires remote control.

Next, the CPU 22 proceeds to S252 and predicts the time when the situation that requires remote control predicted in S242 occurs, the degree of influence on the traffic flow, and the working time required for remote control. The CPU 22 predicts the time when a situation requiring remote control occurs from the travel speed information included in the vehicle information in addition to the travel plan.

Next, the CPU 22 proceeds to S262 and executes grouping in the same manner as in the second embodiment. Finally, the CPU 22 proceeds to S272 and determines the processing order based on the operation policy. For example, as shown in FIG. 14, it is predicted that remote control will be required from time Ta in the region α, then remote control will be required from time Tb in region β, and then remote from time Tc in region γ. Imagine a situation where control is expected to be needed. These predictions were made at time T0. Hereinafter, the work of the operator for remote control in the area α is also referred to as a task α. The regions β and γ are also referred to as tasks β and γ in the same manner.

As for the degree of influence on the traffic flow, the region γ has the largest effect, followed by the region β, and then the region α. The degree of influence on the traffic flow is calculated by statistically processing (for example, integrating) the degree of influence predicted for each of the autonomous driving vehicle 800 and the manually driven vehicle 900 in S232 and S252. In such a situation, consider a case where one operator 1 is in charge of tasks α, β, and γ.

In this embodiment, the cost calculation shown in FIG. 15 is performed, and the work is assigned in the order of the minimum cost. Specifically, as shown in FIG. 15, the cost of all tasks is set as the cost of the task, which is the value obtained by multiplying the delay time from the occurrence of a task to the start by the priority of the task. Accumulate. The higher the priority value, the higher the priority. For example, the priority of task α is 1, the priority of task β is 2, and the priority of task γ is 3. Assuming that the tasks are performed in the order of task occurrence, task α, β, and γ, the cost of task α is 1 minute × 1 = 1, the cost of task β is 1 minute × 2 = 2, and the task is task. Since the cost of γ is 7 minutes x 3 = 21, the total is 24.

When the same cost calculation is carried out for all the ways, when α, γ, and β are carried out in this order, the total cost value is 17 which is the minimum. Therefore, as shown in FIG. 14, the CPU 22 decides to carry out in the order of α, γ, β.

According to the present embodiment described above, the influence on the traffic flow can be further reduced.

The fourth embodiment will be described. The description of the fourth embodiment mainly focuses on the differences from the third embodiment. The points not particularly described are the same as those in the third embodiment.

In the fourth embodiment, as shown in FIG. 16, the content of the allocation process is different from that of the third embodiment. When the processing order determination process is completed, the CPU 22 proceeds to S303 to determine the number of people to be allocated for each group. S303 will be described with reference to FIG. 17 as an example.

FIG. 17 shows a situation in which traffic control by hand signals is carried out in the area X as in FIG. However, in FIG. 17, all the vehicles affected by the traffic control are self-driving cars 800. In such a situation, the CPU 22 in the present embodiment determines the number of people to be allocated to be less than seven even if seven operators can be assigned. Specifically, it will be decided to be two people.

After that, the CPU 22 proceeds to S313 and determines whether or not there is a group in which the allocated number of people is insufficient. If there is a group in which the number of people to be allocated is insufficient, the CPU 22 determines YES in S313, and executes S320 to S340 as in the third embodiment. If there is no group with an insufficient number of people to be allocated, the CPU 22 determines NO in S313 and ends the allocation process. In the situation shown in FIG. 17, if two operators are assigned, the allocation process ends.

Changes in allocation within the group will be described with reference to FIGS. 17 and 18. 17 and 18 show the self-driving cars 800A to 800G. In FIG. 17, 1 is added to the end of the code of the self-driving car 800A. This 1 indicates that the operator 1 is assigned to the self-driving car 800A. Similarly, the self-driving car 800B2 indicates that the operator 2 is assigned to the self-driving car 800B. This allocation is executed by the CPU 22.

The self-driving car 800A1 is traveling in the opposite lane in the situation shown in FIG. The remote control is started when the self-driving car 800A1 is located in front of the traffic control staff. The operator 1 executes remote monitoring of the automatic driving vehicle 800A1 until the automatic driving can be resumed. After moving the self-driving car 800B2 to the front of the traffic control staff, the operator 2 stops the self-driving car 800B2 and confirms the hand signal by the captured image of the front included in the vehicle information of the self-driving car 800B2.

After that, as shown in FIG. 18, the CPU 22 assigns the self-driving car 800E to the operator 1, confirms that the operator 2 permits the self-driving car 800B2 to travel in the opposite lane, and then operates the self-driving car 800C. Assign to 2.

The operator 1 confirms the hand signal by the captured image of the front included in the vehicle information of the autonomous driving vehicle 800E1. While the stop is instructed by the hand signal, the operator 1 continues to stop the self-driving car 800E1.

When the operator 2 permits the self-driving car 800B2 to travel in the opposite lane, the self-driving car 800B travels by remote control without an operator assignment. In order to carry out such remote control, the CPU 22 stores the operation input by the operator 1 for the autonomous driving vehicle 800A1 in the storage device 26. Then, when the operator 2 permits the autonomous driving vehicle 800B2 to travel in the opposite lane, the CPU 22 calls the operation stored in the storage device 26 and transmits the operation to the autonomous driving vehicle 800B. In this way, the autonomous driving vehicle 800B can travel by remote control by traveling on the same route as the autonomous driving vehicle 800A, even if there is no operator assignment.

When the operator 2 permits the self-driving car 800B2 to travel in the opposite lane, the operator 2 performs the same input operation as the self-driving car 800B2 for the self-driving car 800C2. After that, the operator 2 performs the same input operation as that of the self-driving car 800C2 for the self-driving car 800D2. As a result, the self-driving car 800D2 travels as the self-driving car 800D as shown in FIG. The CPU 22 removes the operator 2 from the group assignment.

When the self-driving car 800D passes through the area X, the traffic control officer who has instructed the self-driving car 800E1 to stop permits the vehicle to travel. Upon confirming the permission, the operator 1 executes an operation input for starting the traveling of the autonomous driving vehicle 800E1. The self-driving car 800E1 can pass through the area X simply by going straight if the traffic control staff moves from the roadway to the sidewalk. Therefore, after the start of traveling of the autonomous vehicle 800E1, the operator 1 inputs the end of the remote control to the interface 120. After that, the CPU 22 removes the operator 1 from the allocation of the group, and ends the processing for the area X.

According to this embodiment, it is possible to avoid allocating operators more than necessary. That is, even if operators are assigned to all of the autonomous driving vehicles 800A to G, for example, the operators assigned to the autonomous driving vehicles 800F and 800G are only on standby while the autonomous driving vehicle 800E is stopped. .. According to the present embodiment, such a situation can be avoided, operators can be efficiently assigned, and smooth remote control can be performed.

The fifth embodiment will be described. The description of the fifth embodiment mainly focuses on the differences from the second embodiment. The points not particularly described are the same as those in the second embodiment.

In the present embodiment, in determining the processing order of the group in S271, the self-driving car 800 having the highest priority in the group is specified, and the priority is compared between the groups to change the processing order for each group. decide.

In the present embodiment, the priority of each autonomous vehicle 800 is determined by integrating the following six items. In another embodiment, the priority may be determined in consideration of at least one of the following six items.

The first item is the short connection time when the self-driving car 800 is a means of transportation for connecting to public transportation. The shorter the connection time, the higher the priority. The case of the above-mentioned connecting transportation means is, for example, a case where a person riding on the self-driving car 800 moves to the nearest station on the self-driving car 800 and then transfers to a railroad. By considering this item, the reliability as a so-called last mile transportation will be enhanced.

The second item is the amount of money paid by the passengers of the self-driving car 800. The passenger can pay the amount to the manager of the control device 20 by using his / her own smartphone or the like. The more money you pay, the higher the priority. By considering this item, the service quality of the self-driving car 800 as a shared car is improved.

The third item is the health condition of the passengers. In other words, the worse the health of the passenger, the higher the priority. The health condition of the passenger is determined by the CPU 22 based on the information obtained from various sensors. The various sensors are, for example, imaging results obtained by measuring instruments such as heart rate, body temperature, and respiratory rate, and a camera for in-vehicle imaging provided in an autonomous driving vehicle. The CPU 22 determines the health state by inputting the imaging result into the neural network constructed by machine learning. By considering this item, it is possible to operate in consideration of human life.

The fourth item is the waiting time of the passengers of the self-driving car 800. That is, it is an integrated value of the time when the self-driving car 800 cannot travel due to waiting for remote control. This accumulated value is reset to zero when the passenger changes. By considering this item, the degree of discomfort of the passenger can be reduced. In other embodiments, an average value or the like may be used instead of the integrated value.

The fifth item is the delay of the currently predicted arrival time with respect to the predicted arrival time at the time of boarding. The greater this delay, the higher the priority. By considering this item, it is possible to increase the on-time performance rate.

The sixth item is the degree of danger of the vehicle. Specifically, when a situation such as a collision accident and an airbag is operating is detected, the priority is high. By considering this item, it is possible to operate in consideration of passenger safety.

The correspondence between the embodiment and the claims will be described. The specific part corresponds to S221, the calculation part corresponds to S231, S232, S252 and S260, the determination part corresponds to S272, the allocation part corresponds to S340, and the grouping part corresponds to S262.

The present disclosure is not limited to the embodiments of the present specification, and can be realized by various configurations without departing from the spirit thereof. For example, the technical features in the embodiments corresponding to the technical features in each embodiment described in the column of the outline of the invention are for solving a part or all of the above-mentioned problems, or one of the above-mentioned effects. Can be replaced or combined as appropriate to achieve part or all. If the technical feature is not described as essential in this specification, it may be deleted as appropriate. For example, the following embodiments are exemplified.

In the above embodiment, some or all of the functions and processes realized by the software may be realized by the hardware. In addition, some or all of the functions and processes realized by the hardware may be realized by the software. As the hardware, various circuits such as an integrated circuit, a discrete circuit, or a circuit module combining these circuits may be used.

The communication device included in the control device may separately include one that communicates with the autonomous driving vehicle and the manually driving vehicle and one that communicates with the remote control device.

20 control device, 24 communication device, 120 interface, 800 self-driving car, 830 communication unit

Claims (17)

It is equipped with a communication device (24) that communicates with a plurality of autonomous vehicles (800) equipped with a communication unit (830), and receives commands input to an interface (120) by an operator to remotely control the autonomous vehicle. , A control device (20) that remotely controls the autonomous vehicle using the communication device.
Among the plurality of autonomous vehicles, a specific unit (S221) that identifies a target vehicle that is an autonomous vehicle to be remotely controlled, and a specific unit (S221).
A calculation unit (S231, S232, S252, S260) that calculates the work time and priority required for remote control of the target vehicle based on the traffic conditions around the target vehicle.
A determination unit (S272) that determines the processing order, which is the order in which remote control is executed for the target vehicle, using the calculation result by the calculation unit.
An allocation unit (S340) that allocates remote control work to the operator according to the processing order determined by the determination unit.
A control device equipped with. The communication device acquires a travel plan of the self-driving car from the self-driving car via the communication unit, and obtains a travel plan of the self-driving car.
The calculation unit predicts the time when the autonomous vehicle will be subject to remote control in the future by using the travel plan, so that the working time and priority for the autonomous vehicle to be remotely controlled in the future will be given. Calculate the degree,
Based on the calculation result of the work time by the calculation unit, the determination unit determines the processing order for executing the remote control in the priority of the calculation unit when the work time is predicted to overlap in the plurality of target vehicles. The control device according to claim 1, which is determined by using the calculation result. The control device according to any one of claims 1 to 2, wherein the calculation unit stores a position where remote control has been performed in the past and a working time thereof, and uses the stored information for the calculation. When there are a plurality of target vehicles, a grouping unit (S262) for grouping the plurality of target vehicles into at least one group using the position information of each of the plurality of target vehicles is provided.
The control device according to any one of claims 1 to 3, wherein the determination unit determines the processing order for each group when there are a plurality of the groups. A plurality of the interface and the operator are prepared.
According to claim 4, when there are a plurality of the target vehicles belonging to the same group, the allocation unit allocates a smaller number of the operators than the number of the target vehicles to the remote control work of the plurality of autonomous vehicles. The control device described. When the first vehicle and the second vehicle which is the following vehicle of the first vehicle are included as the autonomous driving vehicles belonging to the same group, the same command is issued to the first vehicle after the command is transmitted to the first vehicle. 2 The control device according to claim 5, which is transmitted to the vehicle. The determination unit identifies the target vehicle having the highest priority in the group, and determines the processing order for each group by comparing the priority of the specified target vehicle between the groups. The control device according to any one of claims 4 to 6. The control device according to any one of claims 1 to 7, wherein the calculation unit calculates the priority based on the degree of influence on the traffic flow. The control device according to claim 8, wherein the calculation unit evaluates the degree of influence on the traffic flow by the number of vehicles that cannot travel until the target vehicle starts traveling by remote control. The calculation unit according to any one of claims 8 to 9, wherein the calculation unit evaluates the degree of influence on the traffic flow based on the waiting time of a vehicle that cannot travel until the target vehicle starts traveling by remote control. Control device. The calculation unit is any one of claims 8 to 10, which is evaluated by the number obtained by subtracting the number of vehicles flowing out from the road from the number of vehicles flowing into the road on which the target vehicle is located. The control device described in. The control device according to any one of claims 1 to 11, wherein the calculation unit calculates the priority based on the connection time when the target vehicle is a means of transportation connected to public transportation. The control device according to any one of claims 1 to 12, wherein the calculation unit calculates the priority according to the amount paid by the passenger of the target vehicle to the manager of the control device. .. The control device according to any one of claims 1 to 13, wherein the calculation unit calculates the priority based on the health condition of the passenger of the target vehicle. The calculation unit is any one of claims 1 to 14 for calculating the priority based on the time when the passenger was on the vehicle among the times when the target vehicle was stopped by waiting for remote control. The control device described in the section. The calculation unit calculates the priority based on the delay of the arrival time at the destination expected at the present time with respect to the arrival time at the destination expected at the time when the passenger of the target vehicle boarded. The control device according to any one of items 1 to 15. The control device according to any one of claims 1 to 16, wherein the calculation unit calculates the priority based on the degree of danger of the target vehicle.

Images